Chemwork

Discussions

1,071 members

Members

Search

Manage

PSV load calculation for gas expantion fire case

Damoon Nasseri
Lead Process Engineer at Foster Wheeler Energy Ltd

Unfollow Damoon

I have an existing Instrument air dryer with a PSV installed to protect

against fire expansion.
This is an existing facility and I am validating the original design, which
is done 20 years ago.
The problem is if I size PSV according to API, the existing PSV is not
sufficient, and I was wondering if I can take credit (positive or negative)
for having desiccant inside the drum and how much, as this is not an
empty vessel.

Top Contributors in this Group

Saeid Rahimi Mofrad

Senior Specialty Process Engineer at Fluor
See all members

Like Comment (6) Share Unfollow Reply Privately 11 months ago

Your group contribution level

Add to Manager's Choice Close Discussion

Congrats! Regularly add great discussions and

comments to stay a Top Contributor.

Comments
6 comments

Top Contributor
Sam Izadpanah
Principal Process Engineer at Jacobs
Sam

Since the vessel is dry and there is no wetted surface, you need to be concerned with gas
expansion PLUS water vapour coming off the desiccant. You can take positive credit for the
volume occupied by desiccant but I think you should try to calculate the amount of water that will
be released from the desiccant at high temperature. And it is a single component so it will
probably come out of the adsorbent at the same time!
Like Reply privately Delete 11 months ago
Saeid Rahimi Mofrad
Senior Specialty Process Engineer at Fluor
Top Contributor
You need to calculate the heat transfer rate inside the vessel through Q = h A (Tw Tg)
Where
h = free convention heat transfer = 0.59 (Gr Pr)^1/4 if (Gr Pr)< 10^9
h = free convention heat transfer = 0.10 (Gr Pr)^1/3 if (Gr Pr)> 10^9

Ads By LinkedIn Members

Chemical Process Software
Engineered for hundreds of industryspecific processes. Try CHEMCAD
free.
Free Rich Dad Workshop
Best Selling Author brings his team
to: Guildford Slough Watford 15-17
Apr
Process Safety Conference
Leading safety event for the chemical
& process industries. Register now.

Latest Activity
Gr and Pr are calculated at the film temperature of (Tw + Tg)/2
A is the total vessel surface area below 7.4 m (basically cylindrical portion). Tw can be considered
as 593C. Tg can be assumed as the water boiling temperature at the relief valve set pressure.

Rohit Sharma and

Abdul fahim Arangadi
joined a group:

The relief load will be W = Q / water latent heat at relief valve set pressure
This ignores the vessel content heating up period and MCp DT and simply assumes that all heat
input is used for water vaporization.
Delete 11 months ago
Ahmad Ehsani
Management Consultant at Oil and Energy Industries Development co. OEID/ Senior
Technical Consultant at LUKOIL
Ahmad

Dont Worry About the size of the PSV. The vessel is full of desicant so there is small free space
for air or gas . Also the water adsorbed by desicant is not too much to make problem in case of
fire.it can be checked by reviewing the data sheet for dryer .
The real vessel volume for PSV sizing in this case is much less than the apparent vesel volume .
Like Reply privately Delete 11 months ago

Chemwork Chemwork
Discussion Forum
8h ago

Randy Lee started a discussion in

Chemwork. Intriguing infographic on
the Bakken Boom Enjoy a great
infographic on the Bakken play from
Hart... Refinerlink: The Burgeoning

Saeid Rahimi Mofrad

Senior Specialty Process Engineer at Fluor
Top Contributor

Bakken Enjoy a great infographic on

the

Business Service

I don't agree with Ahmad that there is no worry about the size of the PSV in this service. As far as
there is a gas inside the system, the gas expansion needs to be considered. And in this
particular case when the gas temperature reaches water boiling temperature, water will vaporize
and over-pressurize the vessel. It is only a bit more complicated as the liquid inside the vessel is
not in contact with vessel metal similar to fire - wetted case.

Bakken play from Hart...

11h ago

Patrick Howley likes a discussion in

Chemwork. Randy Lee

In fact, referring to API equation for unwetted vessels under fire, the gas volume has no effect on
the relief rate. It does not matter if 80% of the system is filled with desiccant or 10%, the fire rate
is function of DT/Dtime for the trapped gas. Therefore, fire case is applicable and the relief valve
should be adequately sized for that. Furthermore, considering the fact that desiccants are porous
the actual gas volume is most probably much more than the volume of those parts of the vessel
where there is no adsorbent (which is anyway irrelevant).

Basics of Refinery Product

Pipelines Ever want to know
exactly how diesel or
gasoline fuel makes its way to the
local service station? This article gives
you a quick understanding of how
products flow from an oil refinery to the
pump. &... more Refinerlink: How

The solution I have propoesd is indeed the basis on which the API fire equation for unwetted
vessels has been developed. For more information, please refer to the paper titled as "The Basis
of API Correlation for Fire Relief of Unwetted Vessels" on my website
http://www.chemwork.org/board.html
Delete 11 months ago
Baher Binesh, M.Sc likes this

Damoon Nasseri
Lead Process Engineer at Foster Wheeler Energy Ltd

Pipelines Ship Refinery Products to the

Pump Ever want to know exactly how
diesel or gasoline fuel makes its way
to the local service station? This article
gives you a quick understanding of
how products flow from an oil refinery
to the pump. ...
1d ago

Thanks for valuable advise

Damoon

Like Reply privately Delete 11 months ago

Paul Frey
Principal Consultant Relief and Flare at Foster Wheeler Energy Ltd

Paul

See all activity

Did you compare the API gas expansion method with the estimated water vaporisation rate as
proposed by Saeid ?
Like Reply privately Delete 10 months ago

About

Feedback

Privacy & Terms

LinkedIn Corp. 2014

Add a comment...
Send me an email for each new comment.

Add Comment

Publishers

Language

Upgrade Your Account